Walid Bendada

WORK HISTORY

EDUCATION

ABOUT WALID BENDADA

I am a Research Scientist at Deezer, a French music streaming service operating in 180 countries. Since January, 2021 I’ve been simultaneously doing a Ph.D at Université Paris Dauphine, under the supervision of Tristan Cazenave and Thomas Bouabça. My research goal is to apply deep reinforcement learning to large scale interactive music recommendations, with a focus on sequential representation and scalability of inference.Before starting my Ph.D, I worked as a Machine Learning Engineer at Deezer for five years. My main focus was transforming a heuristics and tag-based recommender system (RS) into a collaborative filtering-based RS leveraging state of the art models from deep representation learning, sequential decision making and contextual bandits at scale. Problems tackled include automatic playlist generation, user cold start representation and homepage optimization. Solutions were implemented autonomously and led to dozens of AB-tests frequently improving users engagement on the platform.I graduated in 2015 from Ecole Centrale Paris, completing a Master’s degree in Applied Mathematics, and in 2016 from Sorbonne Université, completing a Master’s Degree in Machine Learning and Data Science. I am deeply passionate about my job and always keen to connect with professionals sharing similar interests; feel free to reach out!